284 



with to show the variation of the abdominal bristles (PI. X\''II, Figs. 

 1—3). It is only in certain species in Cricotopiis and Tanytarsus that I 

 have been able to detect strong hairs on the abdomen in addition to the 

 anal tufts. The anal tufts are two conspicuous groups of hairs, sit- 

 uated upon more or less elevated bases, on the dorsal surface of the last 

 segment. In each of these elevated bases is a small ganglion which 

 would seem to indicate that the hairs are sensory in nature. In the 

 aquatic forms of Ccratopogonincc neither thoracic nor anal pscndopuds 

 are present ; but in the terrestrial forms of that group, and also in the 

 larVcT of other Chirononiida:, both are present, and generally well de- 

 veloped. Sometimes the two of each pair are so fused as to present 

 the appearance of a single pseudopod ; while in others, and particularly 

 in the case of the anal pair in some species of Tanypns, they are re- 

 markbly elongated. The apices of at least the anal pair are crowned 

 with two or more rings of retractile hooks which enable the larva to re- 

 tain its hold upon any surface. The thoracic pair in the case of terres- 

 trial Ccratopogonincc have also strong hooks similar to those on the 

 anal pair; but in CJiironomiis and some other genera the thoracic pair 

 has onlv numerous, rather soft, apical hairs. The eleventh segment in 

 certain species of the genus Chironomiis has either one or two pairs of 

 ventral blood-gills. In lohifenis Say, there is only one pair of these 

 gills, which are situated rather higher on the side of the segment than 

 usual, but in many species these organs are very much elongated and 

 situated low, almost or quite on the latero-ventral surface. In addition 

 to these ventral blood-gills, which are, as far as I know, confined to a 

 few species in the genus Chirononuis, there are generally present on the 

 surface of the twelfth segment two pairs of well-developed dorsal 

 blood-gills. These organs, as far as my observation goes, are repre- 

 sented by at least one pair in all cases except the CcratopogoniUiC, 

 though in at least the aquatic forms of the latter they are probably re- 

 tractile. The form of the dorsal blood-gills varies considerably in the 

 different genera, and even in the different species within a genus. 



The only exception to the foregoing description of the aquatic 

 larvae is to be found in the case of Palpomyia and allied genera, in 

 which the entire body is snakelike, and the only protuberances present 

 consist of four pairs of hairs on the last segment (PI. XVII, Fig. 6). 

 These larvae swim with a peculiar twisting, serpentine movement, re- 

 minding one very forcibly of the motions of an eel. 



The larvae of many aquatic species live free in the water, while 

 others form tubular tunnels in the mud where they lie concealed dur- 

 ing the daytime, many of them being found near the surface of the 

 water after dark, supposedly for the purpose of obtaining a supply of 



